Utveckling av en röstanalytisk kalibreringsfunktionalitet och effektivisering av arbetsflödet i klinisk tillämpning

Technological voice analysis tools play a vital role in speech-language pathology, where effective workflows and consistent measurements are essential for reliable clinical assessments. However, many existing tools involve complex steps for extracting audio values and lack calibration features, limiting usability and consistency. This thesis aimed to improve the usability and functionality of such tools. The methodology was applied in the recording and acoustic analysis program VoiceJournal, where two core features were developed: an automated process for obtaining the fundamental frequency threshold, and an interactive calibration feature. Requirements were identified based on feedback from clinical users and examination of existing tool workflows. The features were then implemented by coding them into the existing software and evaluated through internal and external functional and usability tests with licensed speech therapists. The results showed that the threshold implementation reduced the total task time by 58% and decreased the number of manual steps required. Furthermore, the developed calibration feature allowed recordings to be made under controlled and consistent conditions. Feedback from the speech therapists confirmed that the new implementations were effective and well-appreciated. These findings suggest the new improvements can enhance the usability of the voice analysis tools and ensure meaningful comparisons between recording sessions, contributing to more consistent and efficient voice assessments in speech-language pathology.